Alexander Cameron shares his views in "Cutting Off the Air Supply" on Narrowstep, Joost/The Venice Project, and BT's business of DSL backhauling.

To us, this paper was of quite interest both for Alex's industry insight into BT and the UK's DSL economics but also for his viewpoint on Joost and its hurdles.

While we have remained quiet with our viewpoints on Joost during beta-testing, our feelings are much like Mr. Cameron's in that Joost will present bandwidth issues for users and must win some significant content deals before becoming anything as ground breaking as Skype or Kazaa. Then again, were Skype and Kazaa really ground breaking or were they mere pedestals created by the media and large acquisitions?

Round three of this media grandstanding may be upon us with article headlines stating "Internet TV Has Finally Arrived."

Really?

Minus the serious bandwidth issue facing Joost, one would find it very hard to believe that Joost will be the "best of both the internet and TV worlds" unless they can sign on the entertainment or sports industry. If Joost can't produce some significant content agreements in these fields and others, then it becomes a software-based YouTube with a smaller bandwidth bill.

Now onto Alex's paper...

IPTV/VoD Cutting Off The Air Supply

By Alexander Cameron, Managing Director, Digital TX Ltd.

It was a beautiful day when the head of Disney finally came out and admitted piracy (or the "Dark Net" as they call it) is a business model. A bit like a drug addict coming out of rehab, its hard to admit what you don't want to admit. But we still carry on and on with the same old story like a broken record. HD-DVD's copy protection mechanism has been circumvented as was predicted a long time ago. History is repeating itself with LA's faithful declaring that it would never be practical to swap HD movies. Their fluffers presumably forgot to remind them that they said the same about DVD, which didn't work out exactly as they thought it would. Bandwidth and storage gets cheaper, and PC processors get faster.

Narrowstep have also had a pretty rough time of it, due to the mergence of Jeremy Allaire's latest creation, Brightcove. Their Internet TV product signed up nearly 100 niche channels and they happily strolled along to do their IPO. Since then, they've booted out the CEO, lost a lot of customers and presumably have been scratching their heads deciding what to do next when everyone else is flocking to Brightcove, who do it for free and even get the advertisers for you. Hopefully there is a bright spark with a mug of coffee somewhere in their office that will point out their biggest mistake was to develop around Windows Media and not the universally accessible Flash Video. Time to migrate to Flash Video, provide a "freemium" version of their service, and re-model the "premium" incumbent.

Alas, all this is peanuts compared to the talk around the water cooler about The Venice Project, aka "Joost".

You have to hand it to our favourite P2P software makers. "Joost" is another marketing triumph in abstract and cartoon-like naming. Now you can get "Juice" from NPower, and "Joost" from the men who made Kazaa and Skype. The industry has had its feathers rustled and is in bit of a flap. You could argue that a little creativity goes a long way in a world where imagination almost seems to be at a premium these days. The excitement over Internet TV and fullscreen video over the net has surprised many people and driven them into a bed-wetting frenzy.

The story goes something like this. Long ago, 2 very clever people raided a whole load of open source libraries and put them together to make a peer-to-peer distribution mechanism that became their technological magic sauce. Their first invention was a piece of software called Kazaa, which allowed internet users to trade files on their computers. It upset the record industry a great deal, and it was eventually sold to Sharman Networks, who took it like a female canine from the RIAA later on.

The next jaunt was Internet telephony, so they took the same P2P mechanism and used it to do encrypted voice audio inside a very nicely packaged piece of instant messaging software they called Skype. It was a very usable product that saved everyone money and worked nicely through firewalls. They sold it to Ebay for a few billion dollars. Quite why it was worth that much, nobody knows.

Fast forward to today, and the envy and fear they inspire in people has become greater than the products they have built. The next step is to take the same P2P technology and use it to distribute TV and video, so they codenamed it The Venice Project. They even have a competitor this time in the form of Babelgum, who have been quietly rivalling them with the same idea. It's a lot of hype and hysteria over the simple fact that it's the same people who did Skype and Kazaa, and were very successful.

But past success does not guarantee future success. The same magic sauce of putting a software application on top of users downloading from each other instead of a central location has its limits. The Golden Goose may be nearing the end of its shelf life.

Joost uses the XULRunner engine, which for the uninitiated is essentially the Mozilla Foundation's version of Microsoft's XAML. It is a mark-up based language like HTML and allows for rapid application development. Like Firefox, it is a simple barebones product that has a flexible plug-in architecture that allows others to develop their own custom extensions (widgets) for it. The 240Mb of heavily encrypted live video it transfers every hour is MPEG 4 AVC (H.264 or Part 10), encoded between 300-600Kbps and needing a 1Mbps connection run comfortably. Human perception of time means that the network response must be less than 150ms in any audio/video system to appear to be real time.

There's no doubt the technology is very sexy. The video starts up incredibly quickly and the makers' obsession with unobtrusive usability is an enjoyable trademark, meaning as much of the normal TV experience is preserved. Unlike its predecessors however, the delivery infrastructure is not fully decentralised as there needs to be a critical mass of users watching the same content before it can be. Servers in the UK and Holland seed the grid and originate the streams. P2P offsets the traffic load from a central distribution site (e.g. website, data centre) but with video content still needs localised caching due the sheer size of data that needs to be fed through.

However, there are 2 massive stumbling blocks before Joost enslaves the masses. The first is simple – the success of a TV service of any content depends on what's on TV, i.e. the content. Technology is irrelevant as content drives demand. Without demand, there is no incentive to develop technology. Technologists who employ the "build it and they will come" mentality are always surprised when they don't, because there's nothing on you want to watch. You can stare at Joost in nerd-wonder for hours, but after 20mins you are going to switch it off because you are bored.

Much more importantly, it will bring the ISPs war directly to the content owning communities door. Clearly traffic of that size will not work with BT's capacity-based charging approach or ISP download capping. The UK system is hilariously backward and conservative when it comes to enabling the development of advanced added-value services like TV. A movie is going to cost you half your 1GB monthly download cap, if the stream comes in reliably in the first place without being victim to bottlenecking and/or contention.

The problem is that a battle between the two sides has been brewing over time and got worse with the advent of bandwidth that allows us to easily download video. Someone has to pay the bill for it, and in a market that is continually driving down prices to compete (i.e. reducing margins) and suffering higher costs, something eventually has to give. The situation as it stands today is rosy for content owners, such as movie studios and broadcasters. They get to distribute their products essentially for free over the Internet. They may have a data centre bandwidth bill and traditional overheads (e.g. marketing, CRM, payment processing etc), but they don't have to pay for the pipes.

That means an ISP pays, because they provide the connection into a subscriber's home. They lease backhaul capacity from international transit providers and BT at a cost per Mbit, and the bill just keeps getting bigger as they are forced to buy bigger and bigger pipes for their customers, who want to download video and will go elsewhere if it's too slow or too expensive. You don't need to be a genius to see how it's going to end.

The first steps have been taken in the form of traffic shaping, where heavy P2P users are being warned and kicked off networks. They will bounce around providers until they get fed up, and the ISP mindset is that as long as it's someone else's problem they aren't going to work out a way to accommodate them and welcome them back. But what to do about live streams, download services and heavy web services?

There doesn't seem to be an acceptable answer so far, but an emerging strategy is looming on the horizon that recovers backhaul costs, doesn't need localisation and preserves download caps. Ultimately, the only way many companies can see working is a class action against content owners to block TV/video/P2P traffic to and from their data centres and websites. The content owner will then have to pay the ISP a quarterly/yearly access fee for the traffic to be connected, and doing so will mean it will flow freely to a subscriber and not be counted against their download cap/traffic allocation.

Yes, it s evil capitalism at its finest hour. These ex-academic hippies who created the Internet are fed up with squeezing the last drop out of their infrastructure and will be asking us to accept that the Internet was not designed to handle video, and as a consequence we will need to pay to access premium services. The chances are that content owners will also pass on their access fees to ordinary consumers in subscription or direct price increases.

You really have a death wish if you want to be a UK broadband provider. The sheer number of companies going into administration, sales and acquisitions, mergers, consolidations and crazy "free" offers makes it quite clear that the bloodbath is reaching a pretty horrid market conclusion. You can't offer "free" services when your costs are so high. The competition for subscribers is so insane that only the bravest, and arguably, suicidal businesspeople pour money down the black hole of active customer acquisition. The big 5 are set to dominate, as they are the only ones with the stomachs and wallets for the fight.

So let's look at some basic ISP economics.

If you want to provide DSL services in the UK, you can either resell BT Wholesale's network, or put your own equipment in one of their 6000 or so exchanges (called local loop unbundling, or LLU). The point to point copper wire running between someone's house and the exchange, via the green street cabinets (called the sub-loop) is always maintained by BT Wholesale and has to be rented.

The original incarnation of wholesale DSL products in the UK came in 3 flavours – Datastream, IPStream and Videostream. The latter is now defunct, and IPStream is by far the most common of all. Datastream was designed that the ISP buys capacity to a specific exchange, rather than aggregated bandwidth on a BT Central pipe. Bandwidth works out from £57 to £91 per Mbps but means that ISPs need heavy density on each exchange to make it work as a product or you end up with stranded capacity in certain locations.

IPStream is the main product most consumers are using for their broadband service. Adding a new customer to your network means a £40 activation charge or an £11 migration charge if they already have a DSL service and are changing provider. If you manage to persuade 10,000 people to switch to you, you have a £100k bill to pay. God forbid they are new customers from your fantastic new fluffy marketing campaign you just blew the year's budget on, as you'll be writing a cheque for £400k.

For each of your subscribers, you have a monthly rental to BT of £8.40 (standard) or £12.40 (premium) for a connection of up to 8mbps dependent on line characteristics. This is the charge for connecting from the end users location back to the local exchange. The service is not contended per se but rather has a complicated sharing algorithm that means under busy conditions the maximum throughput could be 400k. "Premium" comes from the old 50:1 20:1 contention ratios used by BT's Home/Office products and supposedly means better performance (e.g. 800k vs. 400k upload, 'preferential' treatment at times of low capacity). So far with our 10,000 subscribers, we are clocking up a minimum monthly rental bill of £84k or £124k.

Now the fun part – backhauling data from the local exchange to an ISP's own network, where it can be routed onto the public internet. Unless you have your own national network, you'll be using BT again. For this, you need a BT Central pipe and some international transit. A Central pipe costs around £31k per month for each 155Mbps capacity, or £200 per Mbit.

No, you read that correctly. £200 per Mbit capacity. No, seriously.

You are allegedly supposed to be able to fit 8000 subscribers onto that (with their individual share of the contented connection of around 30K – no really), but most ISPs generally only manage 4000-5000. With our 10,000 users, we need 2 pipes, so our bill is around £62k a month.

Trunk Internet transit, thankfully, is well priced and well served. Charges for 10Mbps are around £50 at worst case, but with a little negotiating skill and big buying (multiple Gbps) it's entirely possible to get the price down below £9.00. Using those figures, a subscriber watching a 400k stream 24 hours a day costs an ISP around £88.40 per month + VAT. If a subscriber was using their 8Mbps connection at full speed 24 hours a day, they could in theory potentially cost £1600.

BT's 21CN program promises to change the old legacy ATM infrastructure to a unified IMS architecture that bases all the services BT offer on IP. The bandwidth pricing looks like its going to be around £75 per Mbit so ISPs who want to get involved need to build their own national networks to backhaul connections from regional data centres (Points of Prescence, or PoPs). A £125 per Mbit reduction (62% or so) is a step in the right direction but the catch is the estimated £5 million needed to build your own national network.

But of course you could just ignore all this and get your own little LLU train rolling. You have a lot of freedom then, even to do fun things like TV that BT couldn't allow you to. Only one small snag. You need the GDP of a small African nation to do anything remotely sexy. You can unbundle an urban exchange for £15k (hardware and BT co-location fees) and backhaul it with a 1 Gig connection for £30-60k. Then you need to do the maths and geography, which tell you to do that for the top 1000 exchanges, you're going to need something in the range of £80 million pounds. If you give each customer a dedicated 2Mbps line, you're only going to fit 500,000 people on your network, which isn't exactly your investors' or CFO's idea of value for money.

We all deserve better than this. We need more bandwidth – in fact, so much that we're swimming in it. France seems to manage it somehow even with an incumbent monopolist telecoms provider. The economics as they stand today are utterly useless for doing anything other than web browsing. ISPs are doomed, and BT Wholesale has no interest in helping BT Retail's competition other than not being broken up by Ofcom. One day the regulators are going to wake up and realise that we're not as profoundly comatose or stupid enough to believe that they will be in anything other than cahoots with each other until Armageddon. You have to be naïve to think they could act independently of each other. A slap on the wrist doesn't help the market develop or allow innovation to blossom.

The challenge now for service providers is to understand what their products and services will look like in 5 years time. Microsoft and Siemens are leading the charge to own the home "ecosystem", as ultimately all the devices we own will have connectivity with Internet protocols for the local network and larger external one. Triple Play is a misnomer now because its just the decades-old cable model and offers nothing new or exciting – the race is on to consolidate the chaos in the home. Broadband connections will serve and receive video as well as voice and data, as the BT Home Hub demonstrations. The question is only one of how it is put together and commercially differentiated.

Ofcom's new boss, Ed Richards, recently set out his vision of a digital future for the UK, which was an admirable staking of his leadership credentials. But creating that digital future means taking hard choices, having balls of steel and breaking through the fog of vested interests. Until we see that strength shine through in a unified coalition of will across the board, it will continue to be a long way off.


Digital TX is now offering a great value one-day workshop course on IPTV and Video On-Demand (VoD) specifically for web and media professionals. It can help you get up to speed on the latest technologies, content deals, operators and applications across the world, and offer immense value in identifying both new opportunities and threats for your business and personal career.

If you would like more information, call Alex on 07986 37317, email iptvworkshop@digitaltx.tv or visit www.iptvworkshop.co.uk.

About Digital TX Limited
Formed in early 2004, privately owned and based in London (UK), Digital TX Limited is IPTV/VoD consultants for interactive digital television and broadband media. Some of the keywords you might associate with us are IPTV, Video On-Demand, Triple Play, Broadband Entertainment, Video Over IP, Interactive TV, Network Video Gaming and Telco TV.

http://www.iptvworkshop.co.uk